The Normative Theory of Social Exclusion:
Perspectives from Political Philosophy
Abstract: Reducing social exclusion as an aim of policy is a laudable political goal, but the definition of, and normative justification of policies reducing, social exclusion are unclear. This paper first defines social exclusion as used by academics and policy makers. The paper then uses the capability approach to define social exclusion. It then takes various elements of the definition of social exclusion, particularly concerns with respecting choice, human welfare, and concerns with fairness in society, and relates them to contemporary concerns in political philosophy. In so doing, the paper both helps to refine the idea of social exclusion, and also points to theoretical tensions in that idea. It concludes by linking concern with social exclusion to more recent theories of social or relational equality.
I. Introduction: Social Exclusion is Less than All that is Bad in Society
I will address three questions in this essay. First, what is social exclusion? Secondly, what is it about social exclusion that is bad? Thirdly, given how we answer the previous two questions, what normative theory best explains what is bad about social exclusion?
A problem with the idea of social exclusion is that there seem to be as many definitions of social exclusion as there are people working on social exclusion. As a term which initially emerged more from social policy makers than from academics, it is difficult to find a single unifying attribute to all of the possible categories of social exclusion. This is not to say that more academic debates are anymore clear cut; anyone trying to define say, justice, or rationality, knows that the debates have gone on for centuries, so social exclusion should be no different.
This is all not to say that social exclusion does not have a systematic intuition behind it. Although there is undoubtedly a temptation to find everything that could be bad in a society and place it under the label of exclusion, I think there is more to the idea than this. However, much of the time, unfortunately, the term social exclusion can act as a mere catch-all for any wrongs in society.
In this essay, I will attempt to provide an explicitly normative approach to the definition of social exclusion. In order to do this, I will attempt to find a normative framework (or normative political theory) that meets the concerns of social policy makers in addressing social exclusion. The need for a normative framework to help define social exclusion comes from the need of policy makers to set priorities for policy in a budget with finite resources. When there are multiple societal “bads”, where not all can be addressed, one must have a normative classification to determine which bads are the most deserving of rectification.
I will proceed as follows. In the next section, I will first examine definitions of social exclusion, and the categories that social exclusion can cover. I will show that the concerns of social exclusion are well captured by the theory of capabilities and functionings as developed by Amartya Sen. However, I will use Sen’s framework and not the theory. That is, I will be using functionings as a mere means of classification, not as a theory of the human good or as a metric of equality.
In section III I will then move on to discuss the normative theory of social exclusion. I will introduce the normative theory of social exclusion by discussing a debate between Brian Barry and Julian Le Grand on whether social exclusion is, or should be, based on concerns for welfare or for fairness. This follows from a debate about whether voluntary social exclusion, or social isolation as Barry calls it, should be the concern of social exclusion policy. That is, even though a person has chosen to not be part of society, should policy attempt to override their choices in order to promote that person’s welfare? Thus, social exclusion faces the problem of paternalism, raising the question of what justifies social exclusion as a normative concern.In section IV, after addressing what is normatively bad about social exclusion, I begin to question which extant political theory best explains the concern with social exclusion. I develop an approach to social exclusion in which it is seen as wrong because it violates a larger concern with social equality, seen in the form of opportunities for economic, political and social participation. Opportunities for welfare are considered as important as other opportunities, but the goal is not to equalize welfare. The approach will draw on the emerging “relational” approach to equality. As opposed to equalizing any particular good, the concern here is to establish social conditions in which people are treated as equals, regardless of their place in any given social hierarchy (e.g., income, occupational status, social position, etc.). In this theory, the distribution of goods works so as to encourage equal respect among the members of a society, not to equalize any one set of goods. Section V concludes
II. Social Exclusion, Capabilities and Functionings
A great deal of effort is expended in saying what social exclusion is not. For instance, social exclusion is not, or not only, a matter of poverty, nor is it the existence of income inequality. It is not, or not only, the existence of racial discrimination and being a member of an ethnic minority. It is not, or not only, unemployment. It is not, or not only, lack of education, access to health, or other social or political activities.
What, then, is social exclusion?
II. A. Social Exclusion: Groups, Activities, and Location
Much discussion focuses on groups that are socially excluded: the unemployed, the poor, the black community, those exposed to crime, the elderly, etc. But note that it is difficult to say that being a member of any of these, so called, vulnerable groups, is a sufficient condition to make one socially excluded. For an easy example, some long-term unemployed are so because they are long-term wealthy. Or, many members of the black community would balk at the idea that they are socially excluded. They may not be “mainstream”; but they certainly have feel included socially. It is just that they are members of a group that is not so-called mainstream. But they are included in some group.
Given that social exclusion is usually defined by a combination of factors, it is possible, of course, to be a “member” of a single demographic group only, and to be socially excluded. If you are long term unemployed, that may be the only factor that applies to you. You may subsist by the help of your family, and so not be poor, but you are in fact socially excluded by virtue of your long term unemployment.
Nevertheless, a theme common to most, if not all, definitions of social exclusion is that social exclusion is multidimensional. That is, often, social exclusion will have multiple necessary conditions. So, it is usually the particular type of combination of factors that cause social exclusion. It is for this reason that the SEU refers to a “cycle” of social exclusion (SEU, 2004); it is what used to be called “poverty traps”. Another way of putting the point is that social exclusion will always be defined by some multiplicative effect of a set of variables which represent participation.
For instance, it may be that one is not socially excluded in terms of work skills, but if they are unemployed and unable to travel, they will be socially excluded.
The Social Exclusion Unit recently puts the point thusly:
“Social exclusion is about more than income poverty. It is a short-hand term for what can happen when people or areas face a combination of linked problems, such as unemployment, discrimination, poor skills, low incomes, poor housing, high crime and family breakdown. These problems are linked and mutually reinforcing” (SEU, 2004).
So this is an indication of how the mutually reinforcing nature of social exclusion works.
Note further that they say that social exclusion “can happen when people or areas face a combination of linked problems”. This shows that often social exclusion is often seen as something that happens to groups of people. I feel that this move, while often an understandable short hand, is unfortunate. For it is often that many factors of social exclusion only matter when in combination with other factors. For instance, we include “young people” and “women” are classed as factors that affect social exclusion. But of course, women are more likely to be socially excluded only when there are other mitigating factors.[1]
So, my first point is that we must be careful when we refer to “socially excluded groups”, for we must be careful to be sure which individuals in those groups are actually socially excluded. This is not to say that a policy meant to address social exclusion should not aim at helping groups of people or areas, but that as a matter of social exclusion it is perhaps better to begin by looking at the activities which are considered important for a society, and then finding which types of individuals are excluded from partaking in those activities. This will also enable us to better connect the theory of social exclusion with the concerns of justice as a theory of equality of opportunity.
Fortunately, the language of capabilities is well suited to those attempting to connect political theory with the concern with social exclusion. For, ultimately, the aspect of social exclusion that transport that can ameliorate has to do with certain functionings, or capabilities to function. Primarily, as the SEU transport report emphasizes, what is perhaps most important are the, quite specific, functionings; these are: and I quote: “getting to work, learning, healthcare, food shopping, and other key activities”.
Now, we may question whether these are the categories that are most worth mentioning in terms of social exclusion, but I think it is a good starting set of guidelines. This is particularly so for reasons I shall discuss below.
Meanwhile, a definition of social exclusion which might be described as the academically canonical definition of social exclusion, which is from the introduction to the book Understanding Social Exclusion, a work by LSE’s CASE (Centre for the Analysis of Social Exclusion) and written by Burchardt, Le Grand and Piachaud. If we assume that there is a set of core activities which constitute participation in society, then an individual is socially excluded in case two conditions are met:
1.the individual is not participating for reasons beyond his/her control, and
2.he or she would like to participate” (Burchardt et al., 2002b)
The areas of activity which constitute social exclusion include, according to the same set of authors:
“Consumption: the capacity to purchase goods and services (including income and wealth- SWO).
Production: participation in economically or socially valuable activities Political engagement: involvement in local or national decision making
Social interaction: integration with family, friends, and community” (Burchardt et al, 2002b: 31)
I will show in the next section that this definition is inadequate in that it is not grounded in a political theory which takes the relation between choice and welfarism seriously.2 However, for now let us begin to attempt to show the activities which are important for understanding social exclusion. To foreshadow my later point, let me just point out that notice that when social policy makers discuss social exclusion, they refer to particular activities:educational, health, employment, socializing, shopping, etc. Oddly, the academic definition is more expansive, and as I shall show, unacceptably broad, as it could be simply referring to activities people want to engage in.
Although this definition does constrain the set of activities somewhat, it does seem to follow that the definition allows activities of “participation” which seem far from the concern of the state or even the community. This will be so particularly with the category of “social interaction”. For instance, suppose a young man (call him Alan) is in a wheelchair, and all of his friends are able to get to the pub, drink themselves into oblivion, and have arguments. Alan is able to travel to work, the surgery, hospital, educational opportunities, and vote and attend local political debates, but not to the pub. Now, Alan wants to apply for funding from the local social exclusion fund in order to have his taxi fares subsidized so that he can engage in a social activity he wants to engage in, but is prevented from engaging in for reasons beyond his control: namely, going to the pub with his mates, drink himself into oblivion, and argue.[2]
I hope I do not betray any idiosyncratic political theory when I say that regardless of one’s views of justice, equality, or what is wrong with social exclusion, this case does not seem a great priority for state support. However, on the CASE definition, if an individual wanted to engage in this activity, perhaps if he wanted it badly enough, it would count as a form of social exclusion, and hence be deserving of some form of state action.
Now, in defence of the CASE definition, it is perhaps that they assume that there are a set of activities which are valuable for all, but one will not be considered excluded if he does not wish to participate in that activity. However, it is not clear that the direction of defining what counts as exclusion can go the other way: if someone wants to engage in an activity and are unable to do so, then they are excluded by virtue of not having that want satisfied. This then can make the CASE definition of social exclusion a welfarist or preference satisfaction theory, which potentially subjects the theory to problems that I now discuss.
It is because the sources of welfare are so varied and subjective that welfarism encounters so many difficulties in political theory. Although the argument first appeared in Scanlon (1975), Dworkin (2000) elaborates the problem with attempting to equalize welfare as follows. Imagine that a person is hurt in an accident which leaves them unable to use their legs, and the state offers to provide a high-tech wheelchair to help this person. Now imagine that the person replies, “well thanks, but as I am a keen violin player, I would rather have the amount of money for the wheelchair in cash so that I can purchase a Stradivarius violin. This will increase my welfare far more, so I would prefer that”. Now, the intuition behind this example is that most people do not feel that the state is responsible for the provision of a Stradivarius in order to increase that person’s welfare as a response to their accident. Instead, argues Dworkin, the point is that most feel that we are responsible for providing that person with resources to get on with their life; not to indulge whatever may happen to raise their level of welfare.
I mentioned that the capabilities framework provides an obvious candidate for the measurement of social exclusion. Let me now introduce the capabilities framework.
II. B. Social Exclusion, Capability, Functioning, and Welfare
The framework of capabilities derives from a particular debate in political philosophy, of which I shall provide a potted history now.
In Rawls’ theory of justice, perhaps the most important contribution was to provide an opposition to the then-prevailing utilitarianism. Rawls’ developed the idea of primary goods to describe a set of all-purpose resources that all rational agents could agree upon as a means of furthering their aims, whatever their own individual aims or conception of the good. Rawls says that regardless of one’s conception of the good, any rational agent will want to: develop a sense of justice, advance one’s conception of the good; and to specify and revise one’s plans. To achieve these all-purpose ends we require the all-purpose means of primary goods: the basic liberties (freedom of thought and conscience); freedom of movement and association; free choice of occupation; diverse opportunities in terms of final ends; offices and positions of responsibility; income and wealth; and the social bases of self-respect.
The Rawlsian framework dominated political philosophy, with few modifications, until Dworkin published his seminal articles on equality in 1981 (reprinted in Dworkin, 2000; all citations henceforth will be to the reprint). The main point of Dworkin’s argument is to introduce the idea of responsibility into the theory of justice. This argument is roughly as follows: Rawls says that in terms of the distribution of primary goods, we must prioritize distributing to the worst off in society. However, suppose that society’s resources had been evenly distributed, and we are looking at two individuals, Alan and Bjorn. Alan chooses to go into business and make a reasonable living; Bjorn decides to devote his life to his first love, playing tennis. Unfortunately, Bjorn isn’t a very good tennis player, and loses his entire set of allocated funds; he is now the worst off in society and is given priority for distribution of primary goods. Dworkin’s main point is the following: why should Alan, who has decided to work in a boring job in order to ensure his material well-being, be forced to subsidize Bjorn’s risky, and rather more pleasurable, choice of a career in tennis?
Let us look at another problem: suppose that we thought that the point of equality was to be some kind of equality of welfare. Again let us take Alan and Bjorn. Alan has lived a disciplined life, and has simple tastes as he has learned to live within his means. Bjorn meanwhile develops his taste for champagne and the high life. Needless to say, were we to aim at equalizing welfare, we would have to give far more resources to Bjorn than to Alan, as Bjorn has expensive tastes. However, this seems to offend our intuitions about equality: why should Bjorn’s expensive tastes receive more resources from the state than Alan’s simpler tastes? Similarly, should Alan receive less of any state resources because he has developed a very sparse, or cheap, set of tastes? The intuitive answer to both of these questions seems to be no.